Skip to main content
Science Catalogs
Health Sciences

Modelled drone network could deliver defibrillators ahead of ambulances in Scotland

A modelling study led by the University of Strathclyde suggests that drones carrying automated external defibrillators (AEDs) could reach hundreds of out-of-hospital cardiac arrests each year before an ambulance arrives. The work, published in Reliability Engineering & System Safety, used real Scottish cardiac arrest and ambulance data to test networks in two contrasting health board areas.

Why minutes matter

More than 3,000 people in Scotland experience an out-of-hospital cardiac arrest each year. Earlier studies indicate that the chance of survival drops by roughly 7% to 10% for every minute without intervention, which makes the time taken to get a defibrillator to a patient a central concern for emergency planners.

What the model found

For NHS Greater Glasgow and Clyde, the modelled network would make about 490 deliveries a year in which a drone arrived at least one minute before an ambulance. In NHS Grampian, which includes large rural areas where response is harder, the figure was around 195 cases a year.

The researchers also looked at the share of incidents reached within six minutes:

  • Greater Glasgow and Clyde: 24% of incidents received an ambulance response within six minutes, against 86% for modelled drone delivery, with an average drone response time of 4.01 minutes.
  • Grampian: 32% for recorded ambulance responses, against 77% for modelled drones, with an average drone response time of 4.33 minutes.

The modelling concerned delivery of a defibrillator, not treatment of patients.

Planning for real-world uncertainty

The team, which included the Scottish Ambulance Service and NATO's Centre for Maritime Research and Experimentation, developed a Bayesian learning approach to decide where drones should be based. Rather than simply drawing a travel-distance radius around each base, the method accounts for uncertainty in ambulance response times, Scottish wind conditions, flight times and battery consumption. Geographical constraints and no-fly zones were also included.

The approach gave priority to areas where existing emergency medical service coverage was weaker, while selecting locations that gave drones the greatest probability of reaching patients quickly. Treating coverage as a probabilistic quantity, rather than a fixed range, is the main methodological contribution, and could be relevant to other emergency logistics problems where conditions vary.

Limits and next steps

The authors stress that earlier delivery would not automatically mean earlier treatment. That depends on factors such as whether a bystander is present and able to use the device. Professor Paul Gowens of the Scottish Ambulance Service made the same point, describing bystander use of the defibrillator as the key element in the chain of survival.

Professor Edoardo Patelli, corresponding author and head of Strathclyde's Centre for Intelligent Infrastructure, said drones would complement rather than replace ambulances and paramedics. The results come from simulation, so further evaluation is needed to establish how well such a system would work in practice and whether it would improve patient outcomes. Gowens also noted that clinical expertise should be built into guidance for drone-enabled delivery from the outset, so that systems are practical and safe.

For researchers, the study provides a template for combining reliability engineering with prehospital care data. The open questions are operational and clinical: how bystanders are alerted and guided, how drones perform in live trials, and whether any gain in delivery time translates into better survival.

Source: Defibrillator drones could arrive ahead of ambulances in hundreds of cardiac arrest cases (Medical Xpress, 18 November 2026).

Image: AI-generated illustration.

Working on this topic? Find a journal for your research or join the discussion on Scholars Post.